github/codex
4
0
Fork 0
mirror of https://github.com/openai/codex.git synced 2026-10-19 16:17:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

Preserve unparsed shell wrappers in exec policy (#39588)

Browse Source 
## Why

Reducing a heredoc shell script to its inner executable lets a prefix rule for
that executable apply to the entire wrapper, even though the full script was
not parsed as a plain command.

## What changed

- Fall back to evaluating the complete shell wrapper when plain-command parsing
  fails, including for heredoc scripts.
- Keep these commands sandboxed when only the inner executable is allowed.
- Propose the full wrapper as the exec policy amendment when approval is needed.

## Testing

Added exec policy, Unix escalation, and approval scenario coverage for unparsed
and heredoc shell wrappers.

GitOrigin-RevId: 8f65133acb6b7c638263917e1d9137e45990772c
This commit is contained in:
Dylan Hurd
2026-08-20 05:07:17 +00:00
committed by copyberry
parent 3bebaea8f2
commit 4b450d2f1b
8 changed files with 90 additions and 307 deletions
Download Patch File Download Diff File Expand all files Collapse all files
codex-rs/shell-command/src/bash.rs
-224
View File
@@ -159,26 +159,6 @@ pub(crate) fn parse_shell_lc_literal_commands(command: &[String]) -> Option<Vec<
Some(commands)
}
/// Returns the parsed argv for a single shell command in a here-doc style
/// script (`<<`), as long as the script contains exactly one command node.
pub fn parse_shell_lc_single_command_prefix(command: &[String]) -> Option<Vec<String>> {
let (_, script) = extract_bash_command(command)?;
let tree = try_parse_shell(script)?;
let root = tree.root_node();
if root.has_error() {
return None;
}
if !has_named_descendant_kind(root, "heredoc_redirect") {
return None;
}
if has_named_descendant_kind(root, "file_redirect") {
return None;
}
let command_node = find_single_command_node(root)?;
parse_heredoc_command_words(command_node, script)
}
fn parse_plain_command_from_node(cmd: tree_sitter::Node, src: &str) -> Option<Vec<String>> {
if cmd.kind() != "command" {
return None;
@@ -277,43 +257,6 @@ fn parse_literal_shell_word(node: Node<'_>, src: &str) -> Option<String> {
}
}
fn parse_heredoc_command_words(cmd: Node<'_>, src: &str) -> Option<Vec<String>> {
if cmd.kind() != "command" {
return None;
}
let mut words = Vec::new();
let mut cursor = cmd.walk();
for child in cmd.named_children(&mut cursor) {
match child.kind() {
"command_name" => {
let word_node = child.named_child(0)?;
if !matches!(word_node.kind(), "word" | "number")
|| !is_literal_word_or_number(word_node, src)
{
return None;
}
words.push(word_node.utf8_text(src.as_bytes()).ok()?.to_owned());
}
"word" | "number" => {
if !is_literal_word_or_number(child, src) {
return None;
}
words.push(child.utf8_text(src.as_bytes()).ok()?.to_owned());
}
// Allow heredoc constructs that attach stdin to a single command
// without changing argv matching semantics for the executable
// prefix. Other file redirects may write outside the sandbox and
// must not be collapsed to the executable prefix for execpolicy.
"comment" => {}
kind if is_allowed_heredoc_attachment_kind(kind) => {}
_ => return None,
}
}
if words.is_empty() { None } else { Some(words) }
}
fn is_literal_word_or_number(node: Node<'_>, src: &str) -> bool {
if !matches!(node.kind(), "word" | "number") {
return false;
@@ -330,50 +273,6 @@ fn is_literal_word_or_number(node: Node<'_>, src: &str) -> bool {
})
}
fn is_allowed_heredoc_attachment_kind(kind: &str) -> bool {
matches!(
kind,
"heredoc_body"
| "simple_heredoc_body"
| "heredoc_redirect"
| "herestring_redirect"
| "redirected_statement"
)
}
fn find_single_command_node(root: Node<'_>) -> Option<Node<'_>> {
let mut stack = vec![root];
let mut single_command = None;
while let Some(node) = stack.pop() {
if node.kind() == "command" {
if single_command.is_some() {
return None;
}
single_command = Some(node);
}
let mut cursor = node.walk();
for child in node.named_children(&mut cursor) {
stack.push(child);
}
}
single_command
}
fn has_named_descendant_kind(node: Node<'_>, kind: &str) -> bool {
let mut stack = vec![node];
while let Some(current) = stack.pop() {
if current.kind() == kind {
return true;
}
let mut cursor = current.walk();
for child in current.named_children(&mut cursor) {
stack.push(child);
}
}
false
}
fn parse_double_quoted_string(node: Node, src: &str) -> Option<String> {
if node.kind() != "string" {
return None;
@@ -591,30 +490,6 @@ mod tests {
}
}
#[test]
fn heredoc_prefix_does_not_restore_dynamic_words() {
for script in [
"find . -{delete,print}",
"find . -del*",
r"find . -de\lete",
"cat ~",
"cat ~HOME",
"cat =sh",
"c*",
] {
let command = [
"bash".to_owned(),
"-lc".to_owned(),
format!("{script} <<'EOF'\nEOF"),
];
assert_eq!(
parse_shell_lc_single_command_prefix(&command),
None,
"{script:?}"
);
}
}
#[test]
fn rejects_variable_assignment_prefix() {
assert!(parse_seq("FOO=bar ls").is_none());
@@ -689,103 +564,4 @@ mod tests {
assert!(parse_seq("rg -g\"$(pwd)\" pattern").is_none());
assert!(parse_seq("rg -g\"$(echo '*.py')\" pattern").is_none());
}
#[test]
fn parse_shell_lc_single_command_prefix_supports_heredoc() {
let command = vec![
"zsh".to_string(),
"-lc".to_string(),
"python3 <<'PY'\nprint('hello')\nPY".to_string(),
];
let parsed = parse_shell_lc_single_command_prefix(&command);
assert_eq!(parsed, Some(vec!["python3".to_string()]));
let command_unquoted = vec![
"zsh".to_string(),
"-lc".to_string(),
"python3 << PY\nprint('hello')\nPY".to_string(),
];
let parsed_unquoted = parse_shell_lc_single_command_prefix(&command_unquoted);
assert_eq!(parsed_unquoted, Some(vec!["python3".to_string()]));
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_multi_command_scripts() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
"python3 <<'PY'\nprint('hello')\nPY\necho done".to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_non_heredoc_redirects() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
"echo hello > /tmp/out.txt".to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_heredoc_with_extra_file_redirect() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
"python3 <<'PY' > /tmp/out.txt\nprint('hello')\nPY".to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_heredoc_with_variable_assignment() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
"PATH=/tmp/evil:$PATH cat <<'EOF'\nhello\nEOF".to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_herestring_with_chaining() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
r#"echo hello > /tmp/out.txt && cat /tmp/out.txt"#.to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_herestring_with_substitution() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
r#"python3 <<< "$(rm -rf /)""#.to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_arithmetic_shift_non_heredoc_script() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
"echo $((1<<2))".to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
#[test]
fn parse_shell_lc_single_command_prefix_rejects_heredoc_command_with_word_expansion() {
let command = vec![
"bash".to_string(),
"-lc".to_string(),
"python3 $((1<<2)) <<'PY'\nprint('hello')\nPY".to_string(),
];
assert_eq!(parse_shell_lc_single_command_prefix(&command), None);
}
}